JPH0622299A - Picture coder and decoder - Google Patents

Abstract

PURPOSE:To improve the coding efficiency by applying independent coding or inter-picture prediction coding in one picture direction to a picture at an interval when a picture timing between a signal source picture and a display picture is coincident so as to avoid production of useless information. CONSTITUTION:A picture signal inputted from a picture signal input terminal 1 is imparted to an interlace converter 2, in which a noninterlace signal is converted into an interlace signal. That is, one frame is converted into two fields and 24 frames are converted into 48 fields. In the case of picture signals having different picture number per second, the picture is subject to inter-picture prediction coding independently or in one direction at an interval when time of the two pictures (timing) is coincident, and in the case of pictures of the same content, only coding data of the 1st picture are outputted and only the picture subject to independent or one-side inter-picture prediction coding is outputted from the coder for plural frequency and the picture rate is converted. Thus, pictures in duplicate are not coded.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a high-efficiency coding for efficiently coding a signal with a smaller code amount for recording, transmission and display for processing a digital signal. In particular, the present invention relates to an encoding device and a decoding device between image devices in which the image rates of a signal source image and a display image are different, such as in the case of displaying on a normal television receiver. .

[0002]

2. Description of the Related Art <2-3 Pulldown> When a film image such as a movie is viewed on a normal television, it is necessary to convert it because the image rate is different. To convert a film of 24 frames per second into a television signal of 60 fields per second, it is sufficient to convert two frames of film into 5 fields of a television signal. Also, although the television signal is interlaced, the film is not originally scanned, so
Naturally it is non-interlaced. Therefore, 2 pieces are 5
When converting to a field, one frame is converted into two fields corresponding to one frame, and the other is converted into three fields.
There are duplicate fields. This is shown in FIG. 6, where o is an interlaced odd field and e is an even field. This method is "2-3 pulldown"
This means that one out of every five fields is a repetition of the same field. This kind of processing is realized when converting a film image into an electrical signal, either by simply scanning and then using a semiconductor memory, or by performing frame advance of the film in synchronization with the television signal. it can.

<Bidirectional Interframe Prediction> In high-efficiency coding of moving images, the correlation between frames of image signals is used to predict a frame to be coded with a coded frame, and only a prediction error is generated. There is an inter-frame predictive coding for coding. In recent years, motion-compensated inter-frame prediction, in which an image is moved and predicted in accordance with motion, has become more common. Therefore, there is a method of further improving efficiency by using bidirectional prediction. This has a prediction structure of 2 as shown in Fig. 7.
One frame (P frame) out of several frames is predicted only from one direction (forward direction), and the frame (B frame) between them is predicted from both directions from P frame (forward direction and backward direction). There is a way to In this method, the B frame has a high prediction efficiency and the error does not spread to other parts in the non-cyclic prediction, so that the quantization can be set roughly.
As a result, the efficiency is considerably high, and it is used in the MPEG system which is being standardized in ISO / IEC. The same applies when the frame becomes a field of an interlaced signal.

<Structure of Encoding Device> FIG. 4 shows an example of the structure of an encoding device for bidirectional prediction in a system for encoding a film image as a television signal. Here, it is assumed that the image signal is obtained by simply scanning the film image sequentially (non-interlace) and converting it into an electric signal. The image signal input from the image input 1 is the image rate converter 31,
2-3 pulldown processing is performed. It is conceivable that this processing is built in the telecine device separately from the encoding device, and the encoding device receives a signal as a television signal of 60 fields per second. The signal converted into the television signal of 60 fields per second is temporarily stored in the field memory 3 in order to change the coding order by bidirectional prediction. The P (unidirectional prediction) field is output immediately, and the B (bidirectional prediction) field is output after the next P.

The output of the field memory 3 is the predictive subtractor 4
And the prediction signal input from the predictor 9 is subtracted. The output of the predictive subtractor 4, which is the prediction residual, becomes the data encoded and compressed by the intra-field encoder 5. The rate of the data is fixed by the data buffer 6, the data is output from the data output 7, and is guided to the intra-field decoder 12 in the P field. In the field decoder 12, the compressed data becomes a reproduced prediction residual, and in the adder 11, the prediction signal is added to form a reproduced image signal. The reproduced image signal is stored in the field memory 10, delayed, and then guided to the predictor 9. It should be noted that the field memory 10 stores only the P field used for prediction, and stores two fields. The predictor 9 creates a prediction signal according to the type (P / B) of the field to be predicted and the image order, and supplies the prediction signal to the prediction subtractor 4. The image counter 8 counts the vertical synchronizing signal of the television signal in the period of P field, and outputs the information of field type (P / B) and image order. The field memory 3 and the predictor 9 are controlled by this information.

<Structure of Decoding Device> The structure of the decoding device is shown in FIG. The compressed data input from the data input 30 passes through the data buffer 31 and the intra-field decoder 1
The decoded signal is added at 2, and the prediction signal is added at the adder 11 to form a reproduced image signal. The signal is output from the image output 34 through the switch 33 in the B field. In the P field, the switch 33 is in a crossing state, the reproduced image signal is stored in the field memory 10, and the previous P field image stored in the field memory 10 is output from the image output 34 through the switch 33. .

The reproduced image signal stored in the field memory 10 is given to the predictor 9, which makes a predictive signal according to the type (P / B) of the field to be predicted and the image order, and the adder 11 give. Image order information is input from the image order input 35, and the predictor 9 and the changeover switch 33 are controlled thereby. A film image can be encoded by such an encoding / decoding system.
Duplicate fields generated by -3 pulldown result in duplicate data.

[0008]

When the image of the signal source and the transmission display device have different image rates, such as when a film image is viewed on a normal television receiver, the image rate is converted into a standard television signal and then the transmission display is performed. In some cases, the image rate is converted and in another case, the image rate is converted before being displayed. When high-efficiency coding is applied to transmission, if the image rate is converted to a standard television signal before encoding, there are overlapping fields due to the difference in image rate, and the same image is duplicated.
Since it is encoded once, useless information is generated. On the other hand, when encoding is performed without converting the image rate and the image rate is converted after decoding, information is not wasted, but an image memory is required to convert the image rate on the decoding side.

The present invention has been made by paying attention to the above points, and independently or unidirectionally predictively encodes and decodes images at intervals where the time (timing) of the film image and the normal television image match. It is an object of the present invention to provide a film image encoding device and a decoding device in which a decoder does not require additional memory for image rate conversion by repeatedly displaying only the image on the device and converting the image rate. To do.

[0010]

In order to achieve the above object, firstly, both image sources having different image rates between a signal source image on the encoding device side and a display image on the decoding device side, An image coding apparatus using unidirectional inter-picture prediction, wherein the images are independently coded or one-way inter-picture predictive coded at intervals at which the image timings of the signal source image and the display image match. The present invention provides an image encoding device characterized by the above.

Secondly, there is provided an image decoding device using bidirectional inter-picture prediction between image devices having different image rates of a signal source image on the encoding device side and a display image on the decoding device side. An image decoding apparatus is characterized in that the image is independently decoded or one-way inter-picture predictive decoding is performed at intervals at which the image timings of the signal source image and the display image match. Is.

Thirdly, when encoding an image having a plurality of images of the same content, only the encoded data of the first image of the images of the same content is output, and the encoded data of other images of the same content are The image encoding apparatus described above is characterized by being controlled so as not to output.

Fourthly, the image decoding described in the above is characterized in that only the image that has been independently decoded or one-sided inter-picture predictive decoded is output a plurality of times to convert the image rate. The present invention provides a rectification device.

[0014]

[Function] The images are independently or unidirectionally predictively coded at intervals at which the times (timings) of the two images match, and only the coded data of the first image is output for the images of the same content.
Since the encoded data of other images is not output, useless information does not occur. Independent or unidirectional prediction in the decoding device By outputting only the decoded image multiple times and converting the image rate, the image of independent or unidirectional prediction is stored in the image memory because it is used for inter-picture prediction. Since it is stored, the image rate can be converted just by outputting it.
Therefore, no image memory is needed for image rate conversion. When the conversion is not performed, the image rate is the same as that of the signal source, and outputs of both image rates can be obtained.

[0015]

<Embodiment Coding Apparatus 1> FIG. 1 is a block diagram showing a first embodiment of the coding apparatus of the present invention. The same parts as those in the conventional example of FIG. The coding apparatus of FIG. 4 is different from the coding apparatus of FIG. 4 in that the interlace converter 2 is provided instead of the image rate converter, and the field prediction type (P
The setting of B) and the control based on it are different.

In FIG. 1, the image signal input from the image signal input terminal 1 is converted by the interlace converter 2 into a non-interlaced signal. That is, one frame is converted into two fields, and 24 frames (frames) are converted into 48 fields. The configuration after the field memory 3 is the same as the conventional example, and only the processing content is different. The P field is set to a field that is output twice when it is converted into 60 fields on the decoding side. As a result, it is not necessary to output the B field twice, and the decoding device does not need a memory for image rate conversion.

Such a field structure is as shown in FIG.
P-fields are provided at intervals at which the image timings match. Figure 8 shows the same image at different image rates,
One frame of film is surrounded by a solid line, and a plurality of squares in the frame becomes a field. The upper (o) fields are odd fields, the lower (o) fields are even fields, the numbers are the image numbers, P and B are the field prediction types, and the P fields are shaded. A field indicated by-is a field in which encoded data is not transmitted.

In the state (a) of 48 fields per second, P
It can be seen that the field is every two frames, and even-odd is present every time. The image order in the encoding apparatus of FIG. 1 is that of (a), and the operation of the field memory and the predictor follows that. Here, the even-odd order is replaced every two frames, because this is based on the order of (b) when the image rate is converted to 60 fields per second. The interval of P is always 5 fields in 60 fields per second (b).

<Second Embodiment> FIG. 2 is a block diagram showing a second embodiment of the encoding apparatus of the present invention. The same parts as those in FIG. 1 are designated by the same reference numerals. The second embodiment corresponds to the case where 2-3 pulldown is performed when a film image is converted into a television signal and the encoded signal is already 60 fields per second.

In this case, the signal of 60 fields may be returned to 48 fields and then the encoding process may be carried out as in the first embodiment. However, in the second embodiment, fields existing in duplicate within 60 fields. Is omitted and only the effective 48 fields are encoded. As for the apparatus configuration, the interlace converter 2 is replaced by an image rate converter, and a data controller 21 is added to the encoding apparatus of FIG.
The field type processed is that of 60 fields per second (b) in FIG. This is similar to the conventional example in which the P field interval is set to 5 fields.

The processing operation is the same as that of the conventional example except that the overlapping field indicated by-is special (the field next to the P field) and the data controller 21 is connected except for that. At the timing of the overlapping field, the data controller 21 is disconnected and the data is not supplied to the data buffer and is not transmitted. At this time, the predictive subtractor 4 and the intra-field encoder 5 do not have to operate. The data output in this manner is the same as that in the first embodiment shown in FIG.

In this encoding device, it is necessary to know which field exists in duplicate, and that information is necessary. This information will be obtained by the frame number if the operation of the image rate converter is defined by the frame number of the output image. Therefore, when the image signal is once recorded in the VTR or the like, it is necessary to use the VTR capable of managing the frame number. If this information is not known, it is judged by matching between two fields. If the image counter is reset with this information, the operation of the encoder can be properly performed. Also, the P-field periodically resets the prediction, making that field an independent field.

<Embodiment of Decoding Device> FIG. 3 is a block diagram showing a first embodiment of the decoding device of the present invention. The same parts as those in the conventional example shown in FIG. The decoding apparatus of the conventional example shown in FIG. 5 has a data controller 32 and is different in processing operation. It should be noted that this decoding device corresponds to both the coding devices of the first and second embodiments.

When converting the image into 60 fields per second and outputting the image, the decoding apparatus is basically operated in 60 fields, and the image order is as shown in FIG. 8B, so that the field memory 10 and the predictor are used. The operation of 9 follows it. Here, the operation is the overlapping field (P
Only the next field after the field) is special,
Otherwise, the data controller 32 is connected and decoded as in the conventional example. At the overlapping field timing, the data controller 32 is disconnected, no data is supplied to the intra-field decoder, and no decoding is performed. Instead,
The P field image stored in the field memory 10 is repeatedly output.

On the other hand, for a television receiver of 50 fields per second, the image speed is allowed to be 4% faster,
The image of 48 fields per second is output without conversion. In this case, the decoding device is basically operated at 50 fields per second. The image order is as shown in Figure 8 (c),
The interval of P changes every time between 5 fields and 3 fields. Since the image order is changed to a or b, the order of the image data is changed in the data buffer 31. The operations of the field memory 12 and the predictor 9 follow the image order of (c). The data controller 32 is always connected. Thus, with the same data and the same decoding device, only the processing content of the decoding device is changed and 60 fields per second and 50 fields per second are obtained.
It is possible to output an image in both of the (48) fields.

[0026]

The film image coding apparatus and the decoding apparatus according to the present invention use image signals having different numbers of images per second, and the images are independent or one-sided at intervals at which the times (timings) of the two images match. The inter-picture predictive decoding is performed, and only the coded data of the first picture is output for the same contents of the image, and only the pictures that have been independently or unidirectionally inter-picture predictive-decoded by the encoding device are output multiple times. By performing the conversion of 1, the redundant images are not encoded, so that useless information is not generated and the encoding efficiency is improved.

Next, in order to convert the image rate in the encoding device, it is sufficient to output the images of independent or unidirectional prediction stored in the image memory repeatedly, and therefore, the conversion of the image rate is performed. There is no need for an image memory. This does not increase the scale of the decoding device.

Further, the image converted by the encoding device is increased in the ratio to the whole by repeating the image of the independent or unidirectional prediction. Since the image of independent or unidirectional prediction is higher in image quality than the image of bidirectional prediction, the average image quality is improved.

Further, since 60 fields per second and 50 fields per second can be reproduced with the same encoded data, the same media can be reproduced in different television system areas, and the media can be shared. As explained above,
The film image encoding apparatus and the decoding apparatus of the present invention have extremely excellent practical effects.

[Brief description of drawings]

FIG. 1 is a block diagram showing a first embodiment of a (film) image coding apparatus according to the present invention.

FIG. 2 is a block diagram showing a second embodiment of the (film) image encoding apparatus according to the present invention.

FIG. 3 is a block diagram showing an embodiment of a (film) image decoding apparatus according to the present invention.

FIG. 4 is a block diagram showing a conventional example of a film image encoding device.

FIG. 5 is a block diagram showing a conventional example of a film image decoding apparatus.

FIG. 6 is a diagram showing how a film image is converted into a television image.

[Fig. 7] Fig. 7 is a diagram illustrating a prediction relationship of predictive coding having bidirectional prediction.

FIG. 8 is a diagram showing a structure of an image at each image rate.

[Explanation of symbols]

1 ... Image input, 2 ... Interlace converter, 3 ... Field memory, 4 ... Predictive subtractor, 5 ... Intra-field encoder, 6, 31 ... Data buffer, 7 ... Data output, 8 ...
Image counter, 9 ... Predictor, 10 ... Field memory,
11 ... Adder, 12 ... In-field decoder, 13 ... Image sequence output, 20 ... Image rate converter, 21, 32 ... Data controller, 33 ... Changeover switch, 30 ... Data input terminal, 34 ... Image output, 35 ... Image order input.

Claims (4)

[Claims] 1. An image encoding apparatus using bidirectional inter-picture prediction between image apparatuses having different image rates of a signal source image on the encoding apparatus side and a display image on the decoding apparatus side, wherein: An image coding apparatus, characterized in that images are independently coded or one-way inter-picture predictive coded at intervals at which image timings of a signal source image and the display image match. 2. An image decoding device using bidirectional inter-picture prediction between image devices having different image rates of a signal source image on the encoding device side and a display image on the decoding device side, wherein: An image decoding apparatus, characterized in that the image is independently decoded or one-sided inter-picture predictive decoding is performed at intervals at which the image timings of the signal source image and the display image match. 3. When encoding an image having a plurality of images of the same content, only the encoded data of the first image of the images of the same content is output, and the encoded data of another image of the same content is not output. The image coding apparatus according to claim 1, wherein the image coding apparatus is controlled as described above. 4. The image decoding according to claim 2, wherein only the image that has been independently decoded or one-sided inter-picture predictive decoded is output a plurality of times to convert the image rate. Device.